Here are the best things I read this week, their main takeaways, and the most important chart from each piece:

And the Winner Was – T-bills? Q4 2018 Quarterly Letter by GMO

  • 2018 was the first year since 1994 that U.S. Treasury bills outperformed both the S&P 500 and the Bloomberg Barclays Aggregate bond index. 2018 was also unique in that no asset did really well, unlike a year such as 2008 where even long-term U.S. Treasuries posted double-digit returns.
  • Since 1900, stocks have historically return +8.7% real in expansions and +6.7% real in recessions. Not a huge difference. GMO says the more important distinction is whether or not earnings are beating or disappointing expectations. GMO thinks that the odds of a positive surprise are better than average when expectations are really low, and unfortunately expectations were really high at the beginning of 2018.
  • Stock analysts are extremely optimistic about future U.S. earnings but pessimistic about international stocks. U.S. earnings expectations imply that earnings will grow 3x as quickly as U.S. GDP, while international earnings expectations imply that earnings will grow less than half as quickly as international GDP.
  • “Investing where success requires something never seen before seems a lot harder than investing where business as usual would count as a significant positive surprise.” – Ben Inker

The Article’s Most Important Chart

Dividends and Buybacks – Fact and Fiction by Aswath Damodaran

  • The dollar value of buybacks from S&P 500 companies has been higher than the dollar value of dividends for most of the last decade. Solely looking at dividends is an incomplete picture of a stock’s total yield to investors.
  • U.S. companies have paid out >80% of their net income in dividends or buybacks over the past six years. Dividing the buyback and dividend sum by net income plus research and development presents a less extreme view, with this adjusted ratio averaging 50% over the past few years.
  • The bulk of buybacks come from firms with low to moderate debt ratios, challenging the popular narrative that buybacks are mainly fueled by highly indebted companies.
  • Buybacks have increased in popularity around the world, specifically in Canada and the United Kingdom. Overall though, dividends are still the preferred cash redistribution tool for international companies.

The Article’s Most Important Chart

Rebalancing…Not So Fast by Jon Seed

  • An investor that frequently rebalances is risking their solvency if the asset they rebalance into doesn’t rise. There have been 50+ year periods where stock markets have delivered negative real returns.
  • “Charlie and I believe in operating with many redundant layers of liquidity, and we avoid any sort of obligation that could drain our cash in a material way. That reduces our returns in 99 years out of 100. But we will survive in the 100th while many others fail. And we will sleep well in all 100.” – Warren Buffet

The Article’s Most Important Chart

Strategic Rebalancing by Nicolas Granger, Campbell R. Harvey, Sandy Rattray, and Otto Van Hemert

  • A mechanical rebalancing strategy, where winners are sold and losers are bought, is an active strategy. Rebalancing tends to lead to larger drawdowns when stocks are in a prolonged downtrend and continue falling after rebalanced into.
  • For a U.S. stock and bond portfolio, an initial 60% of capital allocated to stocks in 1927 drifted to a 76% allocation by 1929, a 32% allocation by 1932, and close to a 100% allocation over time as stocks have significantly outperformed bonds over the long-term.
  • The researchers found that adding a 10% allocation to a trend following strategy to a 90% allocation in a monthly rebalanced 60/40 portfolio improved the average drawdown by approximately five percentage points.
  • An alternative to a trend allocation is strategically timing rebalance trades based on trend following signals. The researchers call this strategic rebalancing. The researchers show that the negative convexity introduced by rebalancing is countered by trend signals preventing reallocations into downtrending assets.

The Paper’s Most Important Chart

A Potted History of Korean Finance by Horseman Capital

  • “Korea has a compelling record of being a large player in many of the big market blow ups of the last two decades.” – Russell Clark
  • Korea’s investments in the semiconductor industry helped propel the country after the 1997 Asian financial crisis. Korea then liberalized its credit card market to promote growth after the dot com bust. Credit cards in issue went from 39 million in 1999 to 105 million in 2002.
  • Koreans didn’t play a big role in the mortgage backed security market, but they did heavily bet on the Korean won continuing to strengthen against the U.S. dollar. The won then fell by 70% in 2008.
  • Bitcoin demand in Korea was so strong that there was a “kimchi” premium on Korean exchanges.
  • According to Horseman, South Korea is now the largest single market for equity linked autocallable derivatives. These products essentially bet that stock indices will stay in a range and that volatility will stay low. With so much Korean retail demand for short volatility products, and their track record of being at the scene of blow ups, Horseman thinks that volatility will increase.

The Article’s Most Important Chart

The Height of the Hurdle by Anu Ganti

  • The S&P 500 index is 45% less volatile than its average index constituents. The level of index volatility reduction increases as the correlation between index constituents decreases.

The Article’s Most Important Chart

Down on the Farm by Winton Capital

  • There have been a number of market bubbles in animals and farm products.
  • Merino wool: From 1809 to mid-1810 American farmers went crazy over Spanish Merino sheep. A single Merino lamb that might have gone for $100 in 1807 went for $1000 in 1809. After Spain’s peninsular war in autumn 1810, Spanish exports were less controlled, the supply of Spanish Merino sheep increased, and prices in America crashed.
  • Mulberry trees: In 1832, several U.S. states subsidized their silk producers. Demand for mulberry trees, the natural habitat of silkworms, increased. By 1838 the premiums for mulberry trees disappeared.
  • Tea: British tea traders left China in 1839 and turned to the Indian province of Assam. In 1861, new land ownership laws and the importation of cheap labor into India made it easier to produce tea. British demand for tea shares and tea land parcels increased, but by 1865 the buyers realized that the promised profits were illusory and quickly sold out.
  • Rabbits: During the 1868 Meiji Revolution, Japan’s warrior class – the Samurai – were disbanded and given cash distributions. Some of them bought imported European rabbits. Prices for the rabbits quickly rose in 1872. Tokyo’s governor prohibited rabbit auctions in 1873, but the city’s foreign merchants set up their own unregulated auctions. It wasn’t until a “rabbit tax” of 1 yen per month (when the average monthly wage was 0.67 yen per month) that prices fell.
  • Ostrich feathers: Ostrich feathers became a popular fashion accessory in the 1880s and South African farmers responded by raising more ostrich. In 1913, the population of Little Karoo (a town at the center of the mania) grew from practically nothing to 776,000. By 1914 changing fashions, concerns over animal cruelty, and the advent of the automobile resulted in less demand for ostrich feathers and price of feathers crashed.

Why Do Markets Go Up? by Ehren Stanhope

  • Stock markets are some the greatest compounders of wealth the world has ever seen, yet few people understand fundamentally why stock markets tend to rise over the long-term.
  • There are three main sources of returns for stock investors:
    • Earnings Growth: Since 1871, U.S. stock market earnings have grown by 3.99% per year.
    • Dividends: Reinvesting dividends would have added 4.55% annually to the 3.99% in earnings growth. Dividends, when reinvested, represent a redistribution of capital to firms that may have higher earnings growth rates. Dividend payers tend to be more mature firms. When dividends are paid and reinvested, they’re typically reinvested pro-rata into the entire index, not just back into the mature companies that paid the dividends.
    • Valuation: In 1871, the TTM P/E for the U.S. stock market was 11.1. At the end of 2018 it was 22.3. This means the valuation multiple rose by 0.47% per year, but it didn’t happen in a straight line. Valuation expansion or contraction makes up the difference between earnings growth, reinvested dividends, and total returns.
  • 3.99% in earnings growth + 4.55% in dividend reinvestment + 0.47% in multiple expansion = 9.01% total return for the U.S. stock market from 1871 to 2018.
  • Decomposing total returns doesn’t answer why earnings grow over time. Ehren says there are three main reasons:
    • Inflation: Inflationary forces result from increases in costs – like labor or raw materials – when demand overwhelms supply. Inflation encourages consumption and diminishes the future burden of debt repayments. In contrast, deflation discourages consumption (since $1 saved today is worth more tomorrow) and makes it tougher to service debt.
    • Productivity: Generally results from efficiency gains (getting faster at an existing task) or value creation (innovating a new process). Productivity gains tend to be gradual and have historically grown at 2% per year. Inflation and productivity are semi-linked, because higher costs tend to foster innovation.
    • Demographics: Labor is often the most expensive component of production. According to the BEA, labor “represents more than 60% of the value of economic output.” More workers means more people are earning an income, and more income means more consumption, and more consumption means more economic activity.
  • “There are fundamental trends at play that justify the continued rise of markets for decades to come – demographic trends and productivity gains.”

The Article’s Most Important Chart

The Many Colours of CAPE by Farouk Jivraj and Robert J. Shiller

  • A CAPE of 31.21 is in its 96th percentile compared to historical CAPE readings since 1881.
  • There’s a wide spread between the worst and best real 10-year forward returns that have historically occurred following current CAPE valuations. The worst inflation-adjusted 10-year compound return has been -6.1% and the best has been +5.8%.
  • Proponents of CAPE think it’s smart to smooth earnings over the business cycle but critics point to the metric’s low statistical significance and changes in accounting standards.
  • The main debates over CAPE center on the use of earnings to price as the valuation metric, use of a ten-year average of earnings (inflation-adjusted), and out-of-sample efficacy.
  • GAAP has undergone major changes in the last two decades. The most notable from the Financial Accounting Standards Board have been FAS 115, 142, and 144. FAS 115 required that assets held for trading or were available for sale be required to be marked to a fair market price. FAS 142 and 144 required that any asset value impairments must be marked to market, regardless of whether or not the asset is sold. However, these assets cannot be written back up unless they are sold and recorded as capital gain income. This has resulted in steep write-offs during recessions without the corresponding revaluation.
  • The authors touch on why CAPE shouldn’t be used for valuation-based market timing: “Historically, it can be seen that equity markets have both risen and fallen for extended and non-symmetric periods of time and, as such, CAPE does not have a steady-state level”

The Paper’s Most Important Chart